A. Understanding what the speaker intended to say.
1. May be remembered in terms of meaningful words and in
the phonemes for
the words
2. May be remembered in terms of "phonetic
gestures"
3. May be remembered as a little of both 1 and 2
1. Definition = the smallest sound in a spoken word which
can be used to
distinguish one word from another.
2. If changed, would alter the meaning of the word
3. Defined in terms of their sounds
4. Example: character
= /k/ /a/ /r/ /a/ /c/ /t/ /u/ /r/
C. Phonetic Gestures (Features)
1. Definition = Physical movements of the vocal tract that
accompany the
production of the phoneme
2. Figure
3. No two phonemes have the same phonetic features (each is
produced in a
different way)
A. Consonants:
1. Produced by pushing air through a constricted or closed
vocal tract
2. Three types of features allow for the production and
identification of
consonants
a. Voicing
1) Vocal cord
vibration?
2) Voiced or
Unvoiced: bin - pin
b. Manner of Articulation
1) How much
is air flow stopped?
2) Types
a) Stop
Consonants
i)
Complete stop
ii)
Example: /k/, /p/, /t/
b)
Fricatives
i)
Nasal stop, mouth restricted
ii)
Example: /z/, /s/, /f/
c) Nasal
Consonants
i)
Nasal stop, mouth stop
ii)
Example: /m/, /n/
c. Place of Articulation
1) Where is
the air flow stopped?
2) Places
a) Lips:
/p/
b)
Teeth: /f/, /th/
c)
Alveolar Ridge: /t/
d)
Palate: /ch/
e)
Velum: /k/
f)
Glottal: /h/
B. Vowels:
1. Produced with an open vocal tract
2. State of the vocal tract determines which vowel is
produced – mainly
the tongue and lips
a. Highest Part
of the Tongue
1) Front: hid
2) Middle:
carry
3) Back: root
b. Height of tongue
1) Low: fat
2) Medium: late
3) High: treat
c. Rounding of Lips
1) Unrounded: he
2) Rounded: who
A. Definition = a diagram which shows what frequencies are
present in a spoken
word and how strong the frequency
components are.
1. Pattern of frequencies and intensities over time
2. Messy looking for natural speech, but clearer if you
examine a
syllable spoken all by itself
3. Some patterns do exist if you look hard enough
1. Seen as a dark horizontal band in a speech spectrogram
which represents
a prominent band of frequencies in the
utterance.
2. A speech sound can be composed of several formants.
3. Formants are numbered from low frequency to high
frequency.
a. The formant with the lowest frequencies is the first
formant
4. The first two formants are the most important ones for
speech
perception.
5. Vowel sounds are generally represented by formants that
look like flat
bars on a spectrogram.
6. Consonant sounds are very variable in appearance.
C. Coarticulation/Parallel Transmission
1. Spoken phonetic segments overlap
a. "dog" vs. "dad" the /d/ sound
changes depending upon
the vowel which
follows it
2. Parts of many phonemes may be simultaneously present
during the
production of a word.
3. Formant transitions are the shift between the vowel and
consonant
4. This makes it difficult to associate particular sounds
with
perceived
phonemes.
1. Speech is special.
2. Humans have a specialized neural structure that allows
them to decode
speech into phonetic gestures.
a. Because we both perceive and produce speech, it
makes
sense to have
one mechanism to handle both
b. Link between vocal tract, tongue, and perception of
speech
3. We are born with this neural structure.
4. Main proponent is Liberman
1. Speech is not special.
2. We learn to understand and perceive speech just like any
other sound.
3. We are not born with a neural structure designed
specifically for
speech perception.
A. Definition = organizing stimuli into groups with the
consequence that
although we can distinguish between
groups we cannot
distinguish items within a group.
B.
We categorize speech perception in a limited way
C.
Almost everyone performs the same on these tasks
A.
Auditory-visual phenomenon that occurs when an observer
1. Hears an audio tape of one phoneme
2. Simultaneously sees a silent video tape of a speaker
saying a
different phoneme
3. The observer reports hearing a completely different
phoneme that is
actually a compromise of the two.
B. Example:
1. What is seen:
"Ga" - place of articulation: back of the
palate
2. What is heard:
"Ba" - place of articulation: lips
3. What is perceived:
a. A compromise between the visual information and the
auditory
information
b. "Da" - place of articulation is between
the back of the
palate (what
is seen) and the lips (what is heard) - it is
the alveolar
ridge
A. Categorical Perception has been demonstrated in infants as
young as 1 month old.
1. Babies distinguish
between phonemes better than adults.
2. Example:
a. Two 't' sounds of Hindi were distinguishable by all 6-8
month olds
b. The two sounds were also distinguishable by 1 year
olds
whose parents
spoke Hindi
c. The sounds were not
distinguishable by 1 year olds
whose parents
did not speak Hindi
B.
Effect of mode of language (i.e. visual vs. auditory language)
1. At 7-10 months, virtually all hearing babies babble.
2. Deaf infants raised by parents whose only form of
linguistic
communication was American Sign Language
babble with their
hands and fingers.